1 This is an appeal against the decision of the adjudicator (“the Adjudicator”) in The Polo/Lauren Company, L.P. v United States Polo Association [2015] SGIPOS 10 (“the IPOS Decision”), in which the Adjudicator dismissed the plaintiff’s opposition to the registration of the defendant’s trade mark. As noted at [12] of the IPOS Decision, this is not the first time the parties have crossed swords; in Polo/Lauren Co LP v United States Polo Association and another action [2002] 1 SLR(R) 129 (“Polo/Lauren 2002”), the plaintiff unsuccessfully applied to set aside the defendant’s registration of its trademark.

2 The plaintiff is the registered proprietor of a family of trade marks which comprise a single polo player in action (“the Polo Player Device”) used either on its own or in conjunction with other words. ¹ These trade marks are registered in a number of countries around the world for a variety of goods and services. ² Of particular significance is Singapore Trade Mark No T9604857H (“the Opposition Mark”) which the plaintiffs registered in Class 9 for “[s]pectacles, spectacle frames, lenses, sunglasses and parts and fittings therefor”: ³

3 The defendant is the governing body of the sport of polo in the United States and has been so since the 1890s. ⁴ It has nonetheless ventured outside its origins as a sports association into the sale of consumer products such as eyewear, luggage and clothing. On 17 October 2012, the defendant applied to register Trade Mark Application No T1215440A (“the Application Mark”) in Class 9 for “[e]yewear; ophthalmic eyewear frames; reading glasses; sunglasses; eyeglass cases and covers; sunvisors (eyewear)”. ⁵ The Application Mark comprises a graphical representation of two overlapping polo players on horseback in a linear perspective (“the Application Device”). The polo players appear to be in motion, with the front player raising his mallet. The Application Device lies next to the text “USPA” (“the Application Text”):

4 The defendant’s application to register the Application Mark was accepted and published in the Trade Marks Journal on 30 November 2012. On 29 January 2013, the plaintiff filed its Notice of Opposition to the defendant’s application. The plaintiff cited four grounds in the Notice of Opposition but ultimately proceeded on two before the Adjudicator – (a) that the Application Mark had been applied in bad faith under s 7(6) of the Trade Marks Act (Cap 332, 2005 Rev Ed) (“the TMA”); and (b) that the Application Mark was similar to the Opposition Mark and was to be registered for goods or services identical with or similar to those for which the Opposition Mark is protected, and there existed a likelihood of confusion on the part of the public under s 8(2)(b) of the TMA. The plaintiff did not appeal against the Adjudicator’s decision in respect of the former. Therefore I shall deal only with the second ground of objection. Section 8(2)(b) of the TMA reads:

5 There was no dispute that the Opposition Mark, which was applied for on 14 May 1996, ⁶ is an “earlier trade mark” under s 8(2)(b) of the TMA. The Adjudicator held that there was “only an extremely low degree of visual similarity between the two marks in this case” and no aural similarity between the Application Mark and the Opposition Mark. ⁷ Nevertheless, she found the marks to be conceptually identical and on the whole, she was of the view the marks were similar, albeit to a very small degree. ⁸ She also found identity between all the goods of the parties. ⁹ Because of the low degree of similarity between the marks and the nature of the goods, which in the Adjudicator’s view would involve a higher degree of care being exercised in their purchase, she found that there was no likelihood of confusion under s 8(2)(b) of the TMA and accordingly dismissed that ground of the plaintiff’s opposition. This forms the sole basis for the plaintiff’s appeal before me. As the defendant does not contest the identity of goods, ¹⁰ I need only consider the similarity of the marks and the likelihood of confusion on the part of the public.

6 A preliminary issue that falls to be determined is the role of an appellate court in proceedings of this nature. The defendant submits that the IPOS Decision should not be disturbed unless there has been a “material error of principle”. ¹¹ It relies on Future Enterprises Pte Ltd v McDonald’s Corp [2007] 2 SLR(R) 845 (“Future Enterprises”) at [5], where the Court of Appeal cited the following holding of Laddie J in SC Prodal 94 SRL v Spirits International NV [2003] EWHC 2756 (Ch) at [19]:

7 The Court of Appeal in Future Enterprises therefore held at [7] that due to the highly subjective nature of the assessment required in s 8(b) of the TMA, “an appellate court should not disturb the findings of fact of a trade mark tribunal unless there is a material error of principle”, an approach which the plaintiff conceded in its oral submissions to apply in the present case. Nevertheless, as I had observed in MediaCorp News Pte Ltd v Astro All Asia Networks plc [2009] 4 SLR(R) 496 (“MediaCorp”) at [26] and [27], the Court of Appeal in Future Enterprises did not appear to have considered the effect of O 87 r 4(2) of the Rules of Court (Cap 322, R 5, 2014 Rev Ed) (“ROC”):

8 Similar observations were made by Chan Seng Onn J in Valentino Globe BV v Pacific Rim Industries Inc [2009] 4 SLR(R) 577 at [9]–[11]. While his decision was upheld on appeal, neither Future Enterprises nor MediaCorp was discussed. Regardless, as was the case in MediaCorp, it does not make a difference as to whether the appeal is to be by way of rehearing given the findings I made below.

9 It is trite that the assessment of the similarity between marks is directed towards substantive similarity, and comprises of three aspects – visual, aural and conceptual similarities: Hai Tong Co (Pte) Ltd v Ventree Singapore Pte Ltd and another and another appeal [2013] 2 SLR 941 (“Hai Tong”) at [40(a)]. The broad principles that guide the court in assessing the likelihood of confusion arising out of the similarity between competing marks and the parties’ services were laid down by the Court of Appeal in Staywell Hospitality Group Pty Ltd v Starwood Hotels & Resorts Worldwide, Inc and another and another appeal [2014] 1 SLR 911 (“Staywell”) at [15]–[20]:

10 Since the defendant accepts that the Application Mark and the Opposition Mark are conceptually identical, ¹² that is also not a matter which requires my consideration.

11 The plaintiff first says that the Adjudicator had erred in finding that the Application Mark had no dominant component (see the IPOS Decision at [79]). ¹³ The factors which the Adjudicator considered to be relevant in assessing the dominance of a component are that as set out in the IPOS Decision at [60]: ¹⁴

(a) The technical distinctiveness of the component: see Staywell at [28].

(b) The “size” of the component: see Hai Tong at [62(d)(ii)] and [62(e)(i)].

(c) The position of the component within the composite mark: see Hai Tong at [62(d)(ii)].

12 The plaintiff argues that the Adjudicator had merely gone through the factors as a “checkbox exercise” in “mechanically” arriving at her finding. ¹⁵ To the extent that it suggests that the Adjudicator did not consider these factors in their totality, I am of the view that there is no basis for this suggestion. It surely cannot be that the Adjudicator’s exercise of running through each factor is in itself evidence of a “checkbox exercise”; indeed, she would have been remiss not to. The Adjudicator was also careful to note, consistent with the debunking of the “words speak louder” principle in Hai Tong at [62(d)], that none of the factors were in themselves conclusive of the dominance of a component. There is no evidence, in my view, that she had placed undue emphasis on any of the factors.

13 The plaintiff submits that the Application Text does not possess a high degree of technical distinctiveness when compared to the Application Device. ¹⁶ In this regard, it sought to distinguish invented words, which the public can be taught to pronounce, from acronyms such as the Application Text which although pronounceable, would not resonate with the public. It also says that the Adjudicator had given insufficient weight to the relative sizes of the Application Device and the Application Text respectively, as the former is “at least two times bigger”. ¹⁷ Further, it says that the Adjudicator had failed to consider distinctiveness in the ordinary and non-technical sense. ¹⁸

14 I first deal with the last of these arguments. The plaintiff seems to suggest in its written submissions that the Adjudicator had failed to consider distinctiveness in the ordinary and non-technical sense in determining the dominant component of the Application Mark. To the extent that this is the case, I am unable to agree. As explained in Staywell at [23], distinctiveness in the ordinary and non-technical sense simply refers to what is outstanding and memorable about the mark. There is no error in the Adjudicator’s failure to consider distinctiveness in the ordinary and non-technical sense at this stage – that is the very aim of determining the dominant component of the Application Mark: see Staywell at [29]. As stated at [11(a)] above, what the court is concerned with at this stage is technical distinctiveness, ie, the capacity of the mark to function as a badge of origin.

15 In this regard, I do not find the distinction which the plaintiff drew between invented words and acronyms, for which no authority was cited in support, to be persuasive. I cannot see why an invented word would have a greater capacity to function as a badge of origin as opposed to an acronym, particularly where both are pronounceable and have no inherent meaning. Neither do I see any error in the Adjudicator’s finding that the Application Device and the Application Text “compare[d] well” in size, certainly not to any extent that would render her finding as to the lack of a dominant component in the Application Mark to be erroneous. While the Application Text may have been smaller than the Application Device, as the Adjudicator notes at [71] of the IPOS Decision, it was in a highly prominent location, lying adjacent to the Application Device, and was of a size sufficient to stand out on its own. For these reasons, I am similarly of the view that there are no dominant components in the Application Mark.

16 The plaintiff says that even if there were no dominant component in the Application Mark, the Adjudicator was wrong to have found that there was only an extremely low degree of visual similarity. ¹⁹ This is because it was premised on her finding that there was only a small degree of similarity between the Application Device and the Opposition Mark, which the plaintiff submits to be erroneous. The plaintiff’s argument is that because the Application Device also comprises of a polo player on horseback with his mallet raised, it has adopted the whole of the Opposition Mark. ²⁰

17 But it is clear from cases such as Clinique Laboratories, LLC v Clinique Suisse Pte Ltd and another [2010] 4 SLR 510 (“CLINIQUE” versus “CLINIQUE SUISSE”) and Ozone Community Corp v Advance Magazine Publishers Inc [2010] 2 SLR 459 (“GLAMOUR” versus “HYSTERIC GLAMOUR”) that the test for visual similarity is not one of substantial reproduction. Rather, visual similarity is assessed by reference to the overall impressions created by the mark or signs, bearing in mind their distinctive and dominant components: see Hai Tong at [62(a)]. I would point out that even if I were to accept the plaintiff’s submission that the Application Device is highly similar to the Opposition Mark, applying the principle set out in Hai Tong at [62(d)(i)], the Application Text would in any event diminish the overall resemblance between them.

18 Given that I am also of the view that there are no dominant components in the Application Mark, I agree with the Adjudicator that there was only an extremely low degree of visual similarity between the Application Mark and the Opposition Mark. Other than the fact that the Application Device and the Opposition Mark are only slightly similar (the former comprises two polo players on horseback, of which one appears to be in lighter-coloured attire and is holding the mallet facing downwards), the prominence and inherent distinctiveness of the Application Text do significantly reduce the visual similarity between the marks.

19 The plaintiff had conceded at the hearing before the Adjudicator that there was no aural similarity on the ground that the Opposition Mark had no textual element. ²¹ Notwithstanding that, the Adjudicator proceeded to record her observations on this issue (see IPOS Decision at [84]–[86]), and it is these observations that the plaintiff objects to. It submits, as it did in its written submissions to the Adjudicator, that there is aural similarity between the Application Mark and the Opposition Mark. ²² This is because the Opposition Mark would be verbalised as “polo-player”, ²³ while the Application mark would be verbalised as “polo-players”. ²⁴

20 I first note that it is unclear if the Adjudicator’s reasons did in fact constitute findings given that she had stressed on more than one occasion that the plaintiff had not pursued this issue before her. The significance of this depends on the threshold for appellate intervention, which I have discussed above at [6]–[8]: if the test is one of whether the Adjudicator had committed a “material error of principle”, then the plaintiff might be precluded from challenging the Adjudicator’s reasons if her finding as to the lack of aural similarity was made simply on the basis that it was not disputed. However, given my reservations as to whether that is in fact the test to be applied, I nonetheless proceed to consider this issue.

21 In respect of how the Opposition Mark would be verbalised, the plaintiff says that it could only be pronounced as “polo-player” given that it comprises solely of a graphic device of a polo player. The Adjudicator’s view was that the public would likely verbalise the Opposition Mark as “Polo Ralph Lauren mark” given that it was often used in conjunction with the words “POLO RALPH LAUREN”. In this regard, the plaintiff is right to point out that the mark similarity inquiry does not take into account external factors: Staywell at [20]. I therefore respectfully disagree with this aspect of the Adjudicator’s view.

22 The plaintiff refers me to the cases of Dainichiseika Colour & Chemicals Mfg Co Ltd v Office for Harmonisation in the Internal Market (Trade Marks and Designs) (Case T-389/03) [2008] ECR II-58 (“Dainichiseika”) and S Tous, S L v Ng Wee Ping [2010] SGIPOS 6 (“S Tous”) which are cited in the IPOS Decision at [84]. In both cases, it was found that the graphic devices would be verbalised in accordance with what they depicted – in Dainichiseika, a device depicting a pelican was held to be likely to be verbalised as “pelican” while in S Tous, a device comprising of a picture of a bear was found aurally similar to one comprising an outline of a bear. Whether a mark that is found to have no aural component may nonetheless be found to be aurally similar to another mark does not appear to be settled law; for example, in Rovio Entertainment Ltd v Kimanis Food Industries Sdn Bhd [2014] SGIPOS 10, the principal assistant registrar found that there could be no aural similarity given that the mark which the opponent sought to rely on had no aural component. This aspect of the decision was undisputed and was therefore not disturbed in the appeal to the High Court: Rovio Entertainment Ltd v Kimanis Food Industries Sdn Bhd [2015] 5 SLR 618 at [85]. Personally, I have my reservations. To find aural similarity where no aural component exists seems to allow for visual or conceptual similarity to be accounted for within the assessment of aural similarity.

23 But even if I were to accept that the Opposition Mark would be verbalised as “polo-player”, I am not persuaded in respect of the Application Mark. The plaintiff seeks to rely on La Societe Des Brasseries Et Glacieres Internationales v Asia Pacific Breweries Ltd [2006] SGIPOS 5 (“APB”), which involved an application to register a mark which had a graphical depiction of a tiger and the words “BIERE LARUE” within the device. It was held at [35] that the average consumer was more likely to refer to the applicant’s beer as a “tiger” brand beer because the graphical depiction of the tiger was more prominent, and the name of the applicant’s beer was in French and not English. By analogy, the plaintiff says the average Singaporean is more likely to refer to the Application Mark as “polo-players” as the Application Text would not “resonate” with them. ²⁵

24 Putting aside the fact that APB does not bind me, it is clear that this holding in APB was premised on the graphical device being the dominant component of the mark, the Registrar having found that the words “BIERE LARUE” was “the secondary element in the [a]pplicants’ mark”: APB at [36]. As I have already found above, I am of the view that neither the Application Device nor the Application Text is the dominant component of the Application Mark. That being the case, I did not see any particular reason why the reasonable consumer would choose to pronounce the Application Mark as “polo-players” rather than as “USPA”, which while invented, is not unpronounceable. As counsel for the defendant highlighted in his oral submissions, no one has sought to pronounce Nike’s “Swoosh” device even though it is only accompanied by the text “NIKE”, which the public may not know to be a Greek goddess.

25 In Sarika Connoisseur Cafe Pte Ltd v Ferrero SpA [2013] 1 SLR 531 (“Sarika”) at [36], the Court of Appeal stated:

26 The Court of Appeal was also at pains to point out in Sarika at [20] that the distinctiveness of a registered trade mark is not another aspect of, or an element in, the determination of similarity. Rather, it is “a factor to be considered in the visual, aural and conceptual analysis to determine whether the allegedly infringing sign and the trade mark are similar”. It is presumably in reliance of this aspect of the decision in Sarika that the plaintiff submits that the Opposition Mark has acquired technical distinctiveness. ²⁶ As I understand it, its argument is that the Opposition Mark has acquired a greater level of technical distinctiveness that increases the threshold for the Application Mark to be found dissimilar.

27 At first blush, the plaintiff’s approach seems perfectly consonant with Staywell at [24]–[25], which suggests the acquired distinctiveness of a mark plays an integral role in the assessment of similarity between marks:

28 I note at the outset that the factoring of acquired distinctiveness at the marks-similarity stage appears to be at odds with the Court of Appeal’s finding in Staywell at [20] that “the assessment of marks similarity is mark-for-mark without consideration of any external matter” [emphasis added]. Rather, the effect of acquired distinctiveness should be left for the confusion stage of the inquiry, when the court assesses “the effect of objective similarity between the marks, on the perception of consumers” [original emphasis removed; emphasis added in italics]. It is clear that Staywell contemplates the factoring of acquired distinctiveness for trade marks which may not be inherently distinctive and have come to be so under s 7(2) of the TMA. It is also clear that Staywell contemplates that earlier trade marks may have varying degrees of technical distinctiveness. It is less clear whether Staywell contemplates that earlier trade marks which are already inherently distinctive can become more distinctive through use and that this enhanced distinctiveness can be taken into consideration at the marks-similarity stage. In Canon Kabushiki Kaisha v Metro-Goldwyn-Mayer Inc [1999] RPC 117 (“Canon Kabushiki”), the following question was referred to the Court of Justice of the European Union (“the ECJ”) (at [11]):

29 As observed in para 6.38 of Christopher Morcom, Ashley Roughton and Thomas St Quintin, The Modern Law of Trade Marks (LexisNexis, 2012 Ed), “[t]he form of the question was such as might suggest that the German court considered that reputation could have some bearing on the issue of similarity of goods or services, an idea which would seem surprising” [emphasis added]. The question as understood in that manner was not addressed; the ECJ instead treated the question as whether the distinctive character of the earlier trade mark, and in particular its reputation, must be taken into account when determining whether the similarity between the goods or services covered by the two trade marks is sufficient to give rise to the likelihood of confusion: Canon Kabushiki at [12]. Nevertheless, given that this approach was not challenged by either of the parties, I am prepared to consider the plaintiff’s argument that the distinctiveness of an earlier trade mark can be enhanced and assessed on a sliding scale.

30 The plaintiff relies on sales invoices, the majority of which have a composite mark comprising the Opposition Mark together with the words “POLO RALPH LAUREN”. ²⁷ However the Adjudicator found that the plaintiff had not proven that the Opposition Mark had acquired a higher degree of technical distinctiveness as none of the invoices featured the Opposition Mark on its own: see the IPOS Decision at [73]. The plaintiff says that the Adjudicator had erred by failing to consider that distinctiveness may be acquired through use in conjunction with another mark: Lonsdale Sports Ltd v Erol [2014] RPC 15 (“Lonsdale”) at [21]–[23]. That is, the fact that the Opposition Mark had been used in conjunction with the words “POLO RALPH LAUREN” did not prevent it from acquiring distinctiveness.

31 Counsel for the defendant submitted at the hearing before me that I was bound by the Court of Appeal’s holding in The Polo/Lauren Co, LP v Shop In Department Store Pte Ltd [2006] 2 SLR(R) 690 at [17], where the word “POLO” was found not to have acquired distinctiveness through use as the appellant there had never used the word mark “POLO” on its own. The parallels to the present case are obvious. Not only does it suggest that acquired distinctiveness may be taken into consideration at the marks-similarity stage (consistent with Sarika and Staywell), but that distinctiveness may only be acquired through use on its own.

32 Even if I were to accept the plaintiff’s submission that the proposition in Lonsdale is good law in Singapore, I agree with the Adjudicator that the evidence adduced by the plaintiff falls short. Counsel for the plaintiff submitted at the hearing before me that the use through which distinctiveness may be acquired must not only be of the mark on its own but must relate to the goods in question, an approach that Sarika at [36] appears to endorse. In this regard, counsel for the plaintiff could not direct me to any invoice for eyewear in which the Opposition Mark was used, on its own or otherwise.

33 On the other hand, counsel for the defendant referred me to the following trade marks comprising a polo player on horseback which were similarly registered in Class 9:

34  The defendant refers me to Polo/Lauren 2002 at [10], in which the High Court observed that “many common objects may represent diverse as well as related or similar interests” – a point which was illustrated by the number of other companies using a graphic device comprising of a polo player on horseback as part of their trade mark. Relying on the authority of Polo/Lauren 2002, the defendant states: ²⁸

35 As I understand it, the defendant’s argument is not that the Opposition Mark completely lacks technical distinctiveness or that it “consist[s] exclusively of signs or indications which have become customary in the current language or in the bona fide and established practices of the trade” under s 7(1)(d) of the TMA; after all, there is no suggestion that the Opposition Mark is invalid. What it appears to be arguing is that the whole of the Opposition Mark intrinsically has a low level of distinctiveness, such that less obvious differences may suffice to distinguish the Application Mark from it. In this regard, Bently and Sherman, Intellectual Property Law (Oxford University Press, 3rd Ed, 2009) at p 866 and 867 refers to the decision of the European Court of First Instance (“the CFI”) in Castellani SpA v Office for Harmonisation in the Internal Market (Trade Marks and Designs) (Case T-149/06) [2007] ECR II-4755 (“Castellani”), where the CFI found the words CASTELLANI (as part of the device mark sought to be registered in German) and CASTELLUCA (which was an earlier trade mark) to be dissimilar. This finding was based on the fact that both words begin with the letters C-A-S-T-E-L-L, which bear a semantic resemblance to the German word “kastell” meaning “castle”. Accordingly, it held that the use of the word ‘castle’ was very common for the class of goods for which registration was sought (wine), and that consumers would pay therefore pay less attention to that prefix.

36 I am not persuaded by the defendant’s argument. It is unclear to me as to how Polo/Lauren 2002 supports the proposition that the Opposition Mark is less distinctive – the point which the judge sought to make through his observation was that the use of a polo player as a motif in that case in itself signified nothing in trade mark law. There is, in my view, insufficient evidence to demonstrate that the Opposition Mark has a lower level of distinctiveness. The mere fact that there are other registered trade marks comprising a polo player on horse-back is not necessarily indicative of the distinctiveness of the Opposition Mark; those marks could either be dissimilar, or do not lead to a likelihood of confusion. To the extent that the defendant is suggesting that the registration of these trade marks is evidence, similar to Castellani, that the use of polo players in graphical devices has become commonplace in respect of eyewear, there is no such evidence before me.

37 I am therefore in agreement with the Adjudicator that while there may be conceptual identity, there is no aural similarity and there is only a low degree of visual similarity. But as the Court of Appeal was at pains to emphasise in Staywell at [18], these are merely “signposts” to the ultimate question of whether the marks are similar.

38 The plaintiff argues that the Adjudicator failed to apply the test laid down by Sir Wilfred Greene MR in Saville Perfumery Ld v June Perfect Ld and W Woolworth & Co Ld (1941) 58 RPC 147 (“June Perfect”) at p 162, ²⁹ which was cited with approval in Hai Tong at [40(d)]:

39 The plaintiff submits that the Adjudicator instead “approached the criteria of visual, aural and conceptual similarities in a formulaic manner”. In my view, there is nothing to substantiate this contention. It is apparent that the Adjudicator was cognisant of the test in June Perfect; [97] of the IPOS Decision explicitly states that “[u]ltimately, the question is whether the two marks, when observed in their totality, are similar rather than dissimilar, and answering this question is inevitably a matter of impression”. Adopting the same approach, I find myself arriving at the same conclusion as the Adjudicator – the marks are similar on the whole, but only to a low degree.

40 The relevant principles to be applied can be found in Staywell, and are concisely summarised by the Adjudicator in the IPOS Decision at [101]:

41 It is the Adjudicator’s findings in respect of the factors set out in [101(d)] of the IPOS Decision that the plaintiff takes issue with. In particular, the Adjudicator held at [110] of the IPOS Decision that “eyewear is a type of goods the purchase of which involves a fairly high degree of care regardless of their price range”, and that pointed away from a likelihood of confusion. She stated three reasons for this holding. First, eyewear, as a type of goods, is not purchased on a regular basis. Second, it is usually bought through salespersons as it is generally sold in optical shops. Third, it is a type of goods that commands a “higher degree of fastidiousness” on the part of the consumer.

42 The plaintiff argues that the Adjudicator had erred in respect of the second and third reasons. It says that the Adjudicator has failed to take into account the notional uses of the Application Mark and the Opposition Mark, such as on sunglasses and reading glasses with fixed prescriptions. ³⁰ Unlike prescription eyewear, they need not be sold through salespersons at optical shops, but can be sold off the shelves at retail outlets. Even with eyewear bought through optical shops, the plaintiff argues that the high degree of care is exercised with regard to the lenses and the adjustment of the nose pad, rather than frame bearing the trade mark. The defendant’s submissions, on the other hand, largely restate the findings of the Adjudicator. ³¹ With particular regard to the third reason given by the Adjudicator in respect of the “higher fastidiousness” of the consumer, the defendants say that the nature of the type of goods, being one that typically requires close visual inspection, means that consumers are more likely to pay attention to the differences that lie between the Application Mark and the Opposition Mark.

43 In this regard, while I accept that the Adjudicator should have considered other forms of eyewear that can be purchased off the shelves, I am not persuaded that a lower degree of care is exercised in respect of this type of goods. As the Adjudicator noted, eyewear is not purchased on a regular basis. This type of goods is highly personal and is not something that would be purchased in a hurry but rather, would involve a high level of attention being paid by the consumer. I do not think that the care exercised would be restricted only to the lenses and the nose pad; given that the eyewear is likely to be purchased on a myriad of factors including not just comfort but also its appearance, any purchase would likely entail a detailed visual inspection of the eyewear.

44 Taking into account the low degree of similarity between the trade marks, particularly in respect of visual similarity, and the greater attention that the average consumer is likely to pay when purchasing eyewear, I find that there exists no likelihood of confusion on the part of the public under s 8(2)(b) of the TMA. I therefore dismiss the appeal with costs to the defendant.